Re: [Gretl-devel] problems importing dataset

On Mon, 21 Jul 2014, Allin Cottrell wrote: >> I think we need a policy decision here. Allin? > > Yes, we do. I suppose we need to screen out "inf" and "nan" in all > combinations of cases. Maybe I'm just being a little pedantic here, but after some reflection, I'd consider "inf" and "nan" as two cases to treat separately. As long as the focus is on reading data from a file, I'm not able to imagine a scenario in which a datafile an econometrician could be possibly interested in could contain plus/minus infinity. On the other hand, I've seen "NaN" (or variants thereof) used as a synonym of "NA": a missing value. I'm not sure myself if this is relevant. Hmmm.